<![CDATA[Amir Pournasserian]]>https://pournasserian.hashnode.devRSS for NodeFri, 13 Nov 2026 19:09:51 GMT60<![CDATA[Validating a vector database built for RAG: four layers of checks, with the code to run them]]>https://pournasserian.hashnode.dev/validating-a-vector-database-built-for-rag-four-layers-of-checks-with-the-code-to-run-themhttps://pournasserian.hashnode.dev/validating-a-vector-database-built-for-rag-four-layers-of-checks-with-the-code-to-run-themThu, 05 Nov 2026 23:39:06 GMTRunnable checks for chunks, embeddings, retrieval and answers, tested against a 536-vector index.

https://pournasserian.com/writing/validating-a-vector-database-for-rag

]]><![CDATA[Decision models against small LLMs: Jev, Kev, Laya and Mammoth-1 on 6,662 decisions]]>https://pournasserian.hashnode.dev/decision-models-against-small-llms-jev-kev-laya-and-mammoth-1-on-6-662-decisionshttps://pournasserian.hashnode.dev/decision-models-against-small-llms-jev-kev-laya-and-mammoth-1-on-6-662-decisionsThu, 05 Nov 2026 23:36:43 GMTFour decision models and five small LLMs, tested on six public datasets with the harness and raw calls published.

https://pournasserian.com/writing/decision-models-vs-small-llms-2026

]]>